text-generation-inference
293b8125
- ROCm: make CK FA2 default instead of Triton (#1924)
Go
Login via GitHub
Home
Pricing
FAQ
Install
Login
via GitHub
Commit
View On
GitHub
Commit
1 year ago
ROCm: make CK FA2 default instead of Triton (#1924) As per title. Triton autotune overhead is prohibitive, as it needs to be done for each different prompt length.
References
#1924 - ROCm: make CK FA2 default instead of Triton
Author
fxmarty
Parents
f871f114
Loading